How do I install the Fpf Sync skill?

Run npx skillmds@latest add dimonier/fpf-sync in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Fpf Sync safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: executes scripts, makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Fpf Sync?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Fpf Sync free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Fpf Sync?

dimonier (@dimonier) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Fpf Sync by dimonier — Agent Skill

Name: Fpf Sync
Rating: 4.2 (5 reviews)
Author: dimonier

Fpf Sync

Sync the fpf-core skill from the canonical FPF-Spec.md monolith. Use when: (1) the FPF specification has been updated (git pull shows new commits), (2) the fpf-core skill references/ are stale or show wrong counts, (3) running scheduled maintenance on LFW skills. Self-contained — bundles decompose_fpf.py + _common.py directly. No dependencies on external skills. Works only with the local skill at `../fpf-core/` (relative to this skill) — NOT the global user skill.

dimonier Updated Sep 10, 2026

File contents

FPF Core Skill Sync

Bounded context: Keeping ../fpf-core/ in sync with the canonical FPF monolith ../../FPF-Spec.md.

Self-contained — all decomposition logic is bundled in scripts/. No external skill dependencies. Requires only Python 3.10+ and git in PATH.

Quick Start

# Check if update is needed (exit code 0=up-to-date, 1=needs update):
python scripts/sync_fpf_core.py --check

# Full sync (git pull + copy + decompose + update SKILL.md):
python scripts/sync_fpf_core.py

# Preview without writing:
python scripts/sync_fpf_core.py --dry-run

# Sync without running git pull (use local state):
python scripts/sync_fpf_core.py --no-git

Workflow

The script performs these steps in order:

git pull in repo root (../..) — fetch latest FPF-Spec.md
Compare hashes — FPF-Spec.md vs skills/fpf-core/assets/FPF-Spec.md. If identical, exit cleanly
Copy monolith — updated FPF-Spec.md → skills/fpf-core/assets/FPF-Spec.md
Decompose — calls bundled decompose_fpf.parse_fpf_spec() to rebuild all references/*.md + INDEX.md from the monolith. Uses idempotent writes — only changed files are overwritten
Update SKILL.md — refreshes line count (e.g. 100K+) and reference counts (291 pattern reference files + INDEX + 34 context sections)
Report — prints counts, warnings, and errors

Scripts

File	Role	Size
`sync_fpf_core.py`	Orchestrator: git pull → diff → copy → decompose → update SKILL.md	7 KB
`decompose_fpf.py`	FPF monolith parser: ToC parsing, `:End`-marker boundary detection, pattern extraction	15 KB
`_common.py`	Shared utilities: reference formatting, YAML frontmatter, idempotent writes, INDEX generation	7 KB
`errors.py`	Error types for pattern extraction	1 KB

Error Handling

Situation	Script behavior
`git pull` fails (network, auth)	Warning — continues with local state
Source `FPF-Spec.md` missing	Error — exits with code 2
Decomposition produces warnings	Printed — exits with code 0 (non-fatal)
Decomposition produces errors	Printed — exits with code 1
No changes detected	Exits cleanly with code 0

SDC Alignment

This skill implements SDC.SyncDiscipline (Spec-Decomposer LPF) for the specific case of the fpf-core skill:

Unidirectional sync (AS.10:4.1): monolith → references. Never the reverse.
Carrier split (AS.4:4.1): agent reads references/, human edits assets/FPF-Spec.md
Full rebuild on every change: copy monolith → decompose → update dispatcher
Idempotent writes: decompose_fpf.py skips unchanged files

Evolution

If the user is dissatisfied with the result or the FPF-Spec structure changes, offer to update this skill's script or its SKILL.md instructions.

---name: fpf-syncdescription: Sync the fpf-core skill from the canonical FPF-Spec.md monolith. Use when: (1) the FPF specification has been updated (git pull shows new commits), (2) the fpf-core skill references/ are stale or show wrong counts, (3) running scheduled maintenance on LFW skills. Self-contained — bundles decompose_fpf.py + _common.py directly. No dependencies on external skills. Works only with the local skill at `../fpf-core/` (relative to this skill) — NOT the global user skill.---# FPF Core Skill Sync**Bounded context:** Keeping `../fpf-core/` in sync with the canonical FPF monolith `../../FPF-Spec.md`.**Self-contained** — all decomposition logic is bundled in `scripts/`. No external skill dependencies. Requires only Python 3.10+ and `git` in PATH.## Quick Start```bash# Check if update is needed (exit code 0=up-to-date, 1=needs update):python scripts/sync_fpf_core.py --check# Full sync (git pull + copy + decompose + update SKILL.md):python scripts/sync_fpf_core.py# Preview without writing:python scripts/sync_fpf_core.py --dry-run# Sync without running git pull (use local state):python scripts/sync_fpf_core.py --no-git```## WorkflowThe script performs these steps in order:1. **`git pull`** in repo root (`../..`) — fetch latest FPF-Spec.md2. **Compare hashes** — `FPF-Spec.md` vs `skills/fpf-core/assets/FPF-Spec.md`. If identical, exit cleanly3. **Copy monolith** — updated `FPF-Spec.md` → `skills/fpf-core/assets/FPF-Spec.md`4. **Decompose** — calls bundled `decompose_fpf.parse_fpf_spec()` to rebuild all `references/*.md` + `INDEX.md` from the monolith. Uses idempotent writes — only changed files are overwritten5. **Update `SKILL.md`** — refreshes line count (e.g. `100K+`) and reference counts (`291 pattern reference files + INDEX + 34 context sections`)6. **Report** — prints counts, warnings, and errors## Scripts| File | Role | Size ||---|---|---|| `sync_fpf_core.py` | Orchestrator: git pull → diff → copy → decompose → update SKILL.md | 7 KB || `decompose_fpf.py` | FPF monolith parser: ToC parsing, `:End`-marker boundary detection, pattern extraction | 15 KB || `_common.py` | Shared utilities: reference formatting, YAML frontmatter, idempotent writes, INDEX generation | 7 KB || `errors.py` | Error types for pattern extraction | 1 KB |## Error Handling| Situation | Script behavior ||---|---|| `git pull` fails (network, auth) | Warning — continues with local state || Source `FPF-Spec.md` missing | Error — exits with code 2 || Decomposition produces warnings | Printed — exits with code 0 (non-fatal) || Decomposition produces errors | Printed — exits with code 1 || No changes detected | Exits cleanly with code 0 |## SDC AlignmentThis skill implements `SDC.SyncDiscipline` (Spec-Decomposer LPF) for the specific case of the `fpf-core` skill:- **Unidirectional sync** (`AS.10:4.1`): monolith → references. Never the reverse.- **Carrier split** (`AS.4:4.1`): agent reads `references/`, human edits `assets/FPF-Spec.md`- **Full rebuild** on every change: copy monolith → decompose → update dispatcher- **Idempotent writes**: `decompose_fpf.py` skips unchanged files## EvolutionIf the user is dissatisfied with the result or the FPF-Spec structure changes, offer to update this skill's script or its SKILL.md instructions.

Frequently asked questions